Additional functional tools for python not covered in the functools library.

pip install functools-extra

A pipe is a function that takes a value and list of functions and calls them in order. So foo(bar(value)) is equivalent to pipe(value, bar, foo). You can use built-in functions like list, special operators from the operator module or custom functions. All type-hints are preserved.

from functools_extra import pipe
from operator import itemgetter

def add_one(x: int) -> int:
     return x + 1

assert pipe(range(3), list, itemgetter(2), add_one) == 3

Or you can use pipe_builder to create a reusable pipe:

from functools_extra import pipe_builder

def add_one(x: int) -> int:
    return x + 1

def double(x: int) -> int:
    return x * 2

add_one_and_double = pipe_builder(add_one, double)
assert add_one_and_double(1) == 4
assert add_one_and_double(2) == 6

A FunctionGroup makes it possible to implement a Protocol with pure functions. If you've ever found yourself creating classes solely to group related functions together to implement protocols, just use a FunctionGroup instead.

Let's consider a scenario where you have a protocol for saving and loading dictionaries, like the DictIO example below:

from typing import Protocol


class DictIO(Protocol):
    def save_dict(self, dict_: dict[str, str], file_name: str) -> None:
        ...

    def load_dict(self, file_name: str) -> dict[str, str]:
        ...

Traditionally, you'd implement this protocol by creating a class:

import json


class JsonDictIO(DictIO):
    def __init__(self, file_name: str):
        self.file_name = file_name

    def save_dict(self, dict_: dict[str, str]) -> None:
        with open(self.file_name, "w") as f:
            json.dump(dict_, f)

    def load_dict(self) -> dict[str, str]:
        with open(self.file_name) as f:
            return json.load(f)

dict_io = JsonDictIO("example.json")

However, you may wonder why you need a class when you're not modifying its state (you're just using it to store common args) or using any dunder methods. You're essentially using the class to group related functions, so why not use a FunctionGroup instead?

from functools_extra import FunctionGroup


json_dict_io = FunctionGroup()

@json_dict_io.register(name="save_dict")
def save_dict(dict_: dict[str, str], file_name: str) -> None:
    with open(file_name, "w") as f:
        json.dump(dict_, f)

@json_dict_io.register
def load_dict(file_name: str) -> dict[str, str]:
    with open(file_name) as f:
        return json.load(f)

dict_io = json_dict_io(file_name="example.json")

This approach encourages a more functional code style and provides all the advantages of pure functions.

This project is using Rye. Check out the project and run

rye sync

to create a virtual environment and install the dependencies. After that you can run

rye run test

to run the tests,

rye run lint

to check the linting,

rye run fix

to format the project with ruff and fix the fixable errors.

This project is licensed under the terms of the MIT license.